|
Activity Number:
|
227
- Bayesian Variable Selection and Shrinkage in Epidemiology Studies
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 30, 2018 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
|
Abstract #330088
|
Presentation
|
|
Title:
|
Multiethnic Joint Analysis of Marginal SNP Effects
|
|
Author(s):
|
David Conti* and Kan Wang and Chris Haiman and Paul Newcombe
|
|
Companies:
|
University of Southern California and University of Southern California and University of Southern California and MRC Biostatistics Unit
|
|
Keywords:
|
Bayesian model selection;
Bayesian hierarchical models;
genetics;
fine-mapping;
multiethnic
|
|
Abstract:
|
To follow up regions identified through GWAS, multiethnic fine-mapping can improve the ability to identify an underlying causal variant by leveraging different linkage disequilibrium(LD) structures across diverse populations. In this context, fixed-effect meta-analysis(FE) remains the most commonly used approach for its ease of interpretation and its ability to pinpoint the causal SNP - especially under the assumption of a common effect across populations. Here, we upon the joint analysis for marginal statistics (JAM) approach that estimates join models from marginal summary statistics while explicitly accounting for the difference in covariance between SNPs across each population. Through simulations we demonstrate that the approach is better for inference of the number of independent signals within a fine-mapping region and identifying the specific causal variants. We demonstrate the application of this method to several known regions for prostate cancer with summary statistics from large consortiums consisting of individuals of European, African and Asian ancestry.
|
Authors who are presenting talks have a * after their name.
